Diagnostic system

Abstract

A diagnosis system (10) acquires motion condition information capable of identifying a motion condition of a robot arm (12). The diagnosis system (10) acquires detection results of a plurality of sensor devices (20) provided to the robot arm (12). The diagnosis system (10) estimates a detection result of a specific sensor device (20) among the plurality of sensor devices (20) at a second time point by inputting the above motion condition information acquired at a first time point and the detection results of the plurality of sensor devices (20) acquired at the first time point to a simulation model. The diagnosis system (10) estimates a state of the specific sensor device (20) by comparing the estimated detection result of the specific sensor device (20) at the second time point with a detection result of the specific sensor device (20) acquired at the second time point.

Description

Technical Field

[0001] This invention relates to a data processing technology, and more particularly to a diagnostic system. Background Technology

[0002] Patent Document 1 below discloses a working device in which an estimation unit estimates the orientation and magnitude of the force detected by the first force detection unit based on the detection result of the second force detection unit, and an anomaly determination unit determines whether at least one of the first force detection unit and the second force detection unit is abnormal by comparing the estimation result of the estimation unit with the detection result of the first force detection unit.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2020-39397 Summary of the Invention

[0006] The problem the invention aims to solve

[0007] The technology disclosed in the aforementioned Patent Document 1 is a technology for estimating the detection result of the second sensor based on the detection result of the first sensor, aiming to improve the estimation accuracy.

[0008] This invention was made based on the inventor's understanding of the above-mentioned problems, and one object of it is to provide a technique for improving the accuracy of estimating the detection results of sensors installed in a device having a movable part that is driven by a driving force.

[0009] Solution for solving the problem

[0010] To address the aforementioned problems, a diagnostic system according to one aspect of the present invention comprises: an operation condition information acquisition unit that acquires operation condition information capable of identifying the operation conditions of a device, the device having a movable part driven by a driving force; a detection result acquisition unit that acquires detection results from a plurality of sensors installed on the device; a detection result estimation unit that estimates the detection result of a specific sensor among the plurality of sensors at a second time point by inputting the operation condition information acquired at a first time point and the detection results of the plurality of sensors acquired at the first time point into a simulation model; and a sensor state estimation unit that estimates the state of a specific sensor by comparing the estimated detection result of the specific sensor at the second time point with the detection result of the specific sensor among the plurality of sensors acquired at the second time point.

[0011] Furthermore, any combination of the above-mentioned constituent elements, or any manner in which the present invention is manifested in an apparatus, method, computer program, or recording medium storing a computer program, is also effective as a mode of the present invention.

[0012] The effects of the invention

[0013] According to the present invention, it is possible to improve the accuracy of estimating the detection results of sensors installed in devices having movable parts that are driven by a driving force. Attached Figure Description

[0014] Figure 1 This is a diagram showing the structure of the diagnostic system according to the first embodiment.

[0015] Figure 2 This is a block diagram illustrating the functional blocks of the sensor device of the first embodiment.

[0016] Figure 3 This is a block diagram showing the functional blocks of the diagnostic device of the first embodiment.

[0017] Figure 4 This is a flowchart illustrating the operation of the diagnostic system according to the first embodiment.

[0018] Figure 5 This is a block diagram illustrating the functional blocks of the diagnostic device according to the second embodiment.

[0019] Figure 6 This is a flowchart illustrating the operation of the diagnostic system according to the second embodiment. Detailed Implementation

[0020] This document outlines the embodiments. To enable remote diagnostics of equipment such as machine tools, the installation of sensors for these devices is constantly evolving. However, for proper sensor diagnostics, it is necessary to accurately determine whether a sensor is in a reliable state. In this embodiment, a technique is proposed whereby, in a diagnostic system that estimates the detection results of sensors installed on a device and uses these estimates to estimate the state of the sensor, the device's operating conditions are considered when estimating the sensor's detection results. This improves the accuracy of both the estimation of the sensor's detection results and the estimation of the sensor's state.

[0021] <First Embodiment>

[0022] Figure 1The structure of the diagnostic system 10 according to the first embodiment is shown. The diagnostic system 10 diagnoses the state of sensors installed on a device (a robotic arm 12 in the first embodiment), the device having movable parts that are driven by driving forces. The robotic arm 12 includes a first movable part 14a, a second movable part 14b, and a third movable part 14c (collectively referred to as "movable part 14"). Each of the plurality of movable parts 14 includes a mechanical element, such as a joint member, that is driven by driving forces such as hydraulic or electrical forces. The robotic arm 12 of the embodiment has three movable parts 14 and performs three-axis movements, but as a variation, the robotic arm 12 may also have six movable parts 14 and perform six-axis movements.

[0023] The robotic arm 12 also includes a first link 16a, a second link 16b, and a third link 16c (collectively referred to as "link 16"). The first link 16a is a link connecting the first movable part 14a and the second movable part 14b. The second link 16b is a link connecting the second movable part 14b and the third movable part 14c. The third link 16c is a link positioned forward of the third movable part 14c.

[0024] The robotic arm control device 18 sends a first control signal to the first movable part 14a to control its movement, based on the posture and action that the robotic arm 12 should take. Additionally, the robotic arm control device 18 sends a second control signal to the second movable part 14b to control its movement. Furthermore, the robotic arm control device 18 sends a third control signal to the third movable part 14c to control its movement. Each of the first, second, and third control signals contains motion condition information that determines the motion conditions of each movable part 14. Motion conditions may include, for example, data specifying or defining the manner of motion of the movable part 14 (e.g., speed, angle, angular velocity, acceleration, motion time, etc.).

[0025] The diagnostic system 10 is a secondary system constructed independently of the primary system related to the movement of the robotic arm 12, and can be subsequently added to the existing primary system. The diagnostic system 10 includes a first sensor device 20a, a second sensor device 20b, a third sensor device 20c, and a diagnostic device 22. When referring to the first sensor device 20a, the second sensor device 20b, and the third sensor device 20c collectively, they are referred to as "sensor device 20".

[0026] Figure 2This is a block diagram illustrating the functional blocks of the sensor device 20 according to the first embodiment. The blocks shown in this specification can be implemented in hardware by components such as a computer processor, CPU, and memory, electronic circuits, and mechanical devices, and in software by computer programs, etc. However, functional blocks implemented through their cooperation are depicted here. Therefore, those skilled in the art will understand that these functional blocks can be implemented in various forms through a combination of hardware and software.

[0027] The sensor device 20 is mounted as a label on the surface of an article (hereinafter also referred to as the "object") having a defined physical structure. The object can be various electronic devices, electrical devices, mechanical devices, components, or finished products. In the first embodiment, multiple sensor devices 20 are provided on the link 16 of the robotic arm 12, but as a variation, at least some of the multiple sensor devices 20 may also be provided on the movable part 14 of the robotic arm 12. The sensor device 20 includes a detection unit 30, a processing unit 32, an environmental power generation unit 34, a power storage unit 36, and an antenna 38.

[0028] The sensor device 20 serves as a sign on the outer surface ( Figure 2 The printed surface displays various information related to the object. Additionally, in the sensor device 20, [the following information is displayed:] ...and... Figure 2 The components corresponding to each functional block shown are integrally arranged in a sheet-like form. "Sheet-like" means that the length of the sensor device 20 in the thickness direction is shorter than either its longitudinal or transverse length. For example, when the longitudinal and transverse lengths of the sensor device 20 are several centimeters, the length in the thickness direction is 5 millimeters or less. Furthermore, it is desirable that the length of the sensor device 20 in the thickness direction is 1 millimeter or less.

[0029] The detection unit 30 is configured to contact or approach an object and is used to measure the state (or physical quantity) related to the object. The detection unit 30 of the first sensor device 20a measures the position of the first sensor device 20a in the robotic arm 12, that is, in the first embodiment, measures the state related to the first link 16a in the robotic arm 12. The detection unit 30 of the second sensor device 20b measures the position of the second sensor device 20b in the robotic arm 12, that is, in the first embodiment, measures the state related to the second link 16b in the robotic arm 12. The detection unit 30 of the third sensor device 20c measures the position of the third sensor device 20c in the robotic arm 12, that is, in the first embodiment, measures the state related to the third link 16c in the robotic arm 12.

[0030] The state related to the object measured by the detection unit 30 can be one or both of the state of the object itself (the state of one or both of the object's interior and surface) and the state of the object's surroundings (in other words, the environment surrounding the object). Furthermore, the state related to the object can be a single type of physical state or quantity, or a combination of multiple types of physical states or quantities. For example, the state related to the object can also be vibration (e.g., triaxial acceleration) and / or temperature. Additionally, the state related to the object can also be the velocity and / or pressure of a fluid flowing within the power transmission path of the object, and can be measured based on the intensity of ultrasonic or radio wave reflections.

[0031] In the first embodiment, the detection unit 30 measures the vibration at the sensor mounting location (or sensor mounting area). The detection unit 30 outputs a signal (also called a "detection signal") based on the measurement result (detection result) to the processing unit 32.

[0032] The processing unit 32 generates information output from the antenna 38 (hereinafter also referred to as "sensor data") based on the measurement results of the detection unit 30, that is, based on the detection signal output from the detection unit 30 in this embodiment. The processing unit 32 may also perform prescribed operations (such as various filtering processes, anomaly diagnosis processing performed by artificial intelligence functions, etc.) based on the detection signal output from the detection unit 30 to generate sensor data containing the results of its operations.

[0033] Antenna 38 serves as an output unit, outputting data based on the measurement results of detector 30, i.e., sensor data generated by processing unit 32 in this embodiment. Antenna 38 can also function as a communication unit, transmitting sensor data to external devices via Wi-Fi, BLE (Bluetooth Low Energy), or NFC (Near Field Communication). In this embodiment, sensor data transmitted from antenna 38 of sensor device 20 is transmitted to diagnostic device 22 via wireless and wired communication networks.

[0034] The environmental power generation unit 34 converts energy present in the environment surrounding the sensor device 20 into electricity (so-called environmental power generation), and supplies the electricity obtained from power generation as power to operate the various functional blocks of the sensor device 20. The environmental power generation unit 34 can also perform known environmental power generation based on at least one of the following: temperature, humidity, radio waves such as Wi-Fi, electromagnetic waves from the surroundings of the sensor device 20 (including radiation, cosmic rays, and electromagnetic noise emitted from electric motors, etc.), vibration, sound (including ultrasound), light (including visible light, infrared light, and ultraviolet light), and the flow of fluids or powders (wind, waves, etc.). Furthermore, the antenna 38 may also include the functions of the environmental power generation unit 34; in this case, the antenna 38 can also perform data communication and environmental power generation in a time-division multiplexing manner.

[0035] The energy storage unit 36 ​​accumulates the electricity generated by the ambient power generation unit 34 and supplies the accumulated power as power to operate the various functional blocks of the sensor device 20. In this embodiment, the detection unit 30, processing unit 32, and antenna 38 of the sensor device 20 can operate based on the power supplied from the ambient power generation unit 34 or using the power supplied from the energy storage unit 36. The energy storage unit 36 ​​can be a capacitor (including an electric double-layer capacitor) or a secondary battery (e.g., a lithium-ion battery, a solid lithium-ion battery, or an air battery).

[0036] return Figure 1 The diagnostic device 22 is an information processing device connected to the first sensor device 20a, the second sensor device 20b, and the third sensor device 20c via a wireless communication network and a wired communication network consisting of access points, switches, routers, etc. (not shown). The diagnostic device 22 performs data processing to diagnose the state of a specific sensor device 20 among the first sensor device 20a, the second sensor device 20b, and the third sensor device 20c. Hereinafter, the specific sensor device 20 to which the state is being diagnosed will be referred to as the "diagnostic target sensor." In the first embodiment, the diagnostic target sensor is the third sensor device 20c.

[0037] Figure 3 This is a block diagram illustrating the functional blocks of the diagnostic device 22 according to the first embodiment. The diagnostic device 22 includes a control unit 40, a storage unit 42, and a communication unit 44. The control unit 40 performs various data processing operations. The storage unit 42 stores data referenced or updated by the control unit 40. The communication unit 44 communicates with external devices according to a predetermined communication protocol. In the first embodiment, the control unit 40 sends and receives data with the robotic arm control device 18, the first sensor device 20a, the second sensor device 20b, and the third sensor device 20c via the communication unit 44.

[0038] Storage unit 42 includes model storage unit 46 and diagnostic information storage unit 48. Model storage unit 46 stores data of a simulation model used to estimate the detection results of the diagnostic target sensor. Details of the simulation model are described later. Diagnostic information storage unit 48 stores diagnostic information representing the estimation results related to the state of the diagnostic target sensor. The diagnostic information may also include information indicating the state of the diagnostic target sensor (e.g., normal or abnormal) as a diagnostic result, and information indicating the date and time when the state of the diagnostic target sensor was diagnosed.

[0039] The control unit 40 includes an operation condition information acquisition unit 50, a detection result acquisition unit 52, a detection result estimation unit 54, a sensor state estimation unit 56, and a diagnostic information provision unit 58. A computer program implementing the functions of these multiple functional blocks can be stored on a specified recording medium or installed into the storage device of the diagnostic device 22 via that recording medium. Alternatively, the computer program can be downloaded and installed into the storage device of the diagnostic device 22 via a communication network. The CPU of the diagnostic device 22 can also read the computer program from the main memory and execute it to perform the functions of each functional block.

[0040] The motion condition information acquisition unit 50 acquires multiple motion condition information (in the first embodiment, a first control signal, a second control signal, and a third control signal) sent from the robotic arm control device 18 to the robotic arm 12, which can identify the motion conditions of multiple movable parts 14. As a variation, the motion condition information acquisition unit 50 may also acquire the above-mentioned multiple motion condition information sent from the robotic arm control device 18 from the robotic arm 12, or it may acquire the above-mentioned multiple motion condition information sent from the robotic arm control device 18 from a relay device (not shown) that relays the communication between the robotic arm control device 18 and the robotic arm 12.

[0041] The detection result acquisition unit 52 acquires the detection results of multiple sensor devices 20 installed on the robotic arm 12. Specifically, the detection result acquisition unit 52 acquires first sensor data representing the detection result of the first sensor device 20a transmitted from the first sensor device 20a, second sensor data representing the detection result of the second sensor device 20b transmitted from the second sensor device 20b, and third sensor data representing the detection result of the third sensor device 20c transmitted from the third sensor device 20c. The first sensor data, second sensor data, and third sensor data in the first embodiment all include information related to vibration at the sensor installation location (e.g., amplitude and frequency).

[0042] Here, the simulation model of the first embodiment is described. The simulation model is a mathematical model that takes as input the motion condition information acquired at a first time point and the detection results (vibration information in the first embodiment) of multiple sensor devices 20 represented by multiple sensor data acquired at the first time point, and estimates the detection results of the diagnostic target sensor determined in advance among the multiple sensor devices 20 at a second time point. The mathematical model can be described as a calculation formula or a function. In addition, the simulation model can also be a so-called digital twin model that simulates the motion of the movable part 14 of the robotic arm 12 and the detection results of the sensor devices 20. In the first embodiment, the second time point is set to be the same as the first time point, but as a variation, the second time point can also be a time point later than the first time point.

[0043] The simulation model of the first embodiment is a mathematical model of the operating conditions input to the plurality of movable parts 14. Alternatively, the simulation model of the first embodiment is a mathematical model of the detection results of the sensor devices 20 (excluding the diagnostic target sensor) input to the plurality of sensor devices 20. For example, the simulation model could also be a regression equation with the operating conditions represented by the first control signal, the operating conditions represented by the second control signal, the operating conditions represented by the third control signal, the detection results of the first sensor device 20a represented by the first sensor data, and the detection results of the second sensor device 20b represented by the second sensor data as explanatory variables, and the detection result of the diagnostic target sensor, i.e., the third sensor device 20c, as the target variable.

[0044] Alternatively, when constructing the simulation model of the first embodiment, the actual values ​​of the following can be collected as sample data: the action conditions represented by the first control signal, the action conditions represented by the second control signal, the action conditions represented by the third control signal, the detection results of the first sensor device 20a represented by the first sensor data, the detection results of the second sensor device 20b represented by the second sensor data, and the detection results of the third sensor device 20c represented by the third sensor data. Furthermore, the coefficients of each explanatory variable can be determined by performing multiple regression analysis based on multiple sample data to generate the simulation model data.

[0045] The detection result estimation unit 54 estimates the detection result of the diagnostic target sensor at a second time point by reading data from the simulation model stored in the model storage unit 46 and inputting the motion condition information acquired at the first time point and the detection results of the multiple sensor devices 20 acquired at the first time point into the simulation model. In the first embodiment, the detection result estimation unit 54 inputs the motion conditions of the multiple movable parts 14 into the simulation model. In addition, the detection result estimation unit 54 inputs the detection results of the sensors other than the diagnostic target sensor among the multiple sensor devices 20 into the simulation model.

[0046] Specifically, the detection result estimation unit 54 inputs the operating conditions represented by the first control signal, the second control signal, and the third control signal acquired at the first time point into the simulation model. Additionally, the detection result estimation unit 54 inputs the detection results of the first sensor device 20a, represented by the first sensor data acquired at the first time point, and the detection results of the second sensor device 20b, represented by the second sensor data, into the simulation model. Furthermore, the detection result estimation unit 54 acquires the detection results (in the first embodiment, vibration-related estimates) of the third sensor device 20c, the diagnostic target sensor, output from the simulation model at the second time point.

[0047] The sensor state estimation unit 56 compares the detection result of the diagnostic target sensor at the second time point estimated by the detection result estimation unit 54 with the detection result of the diagnostic target sensor obtained at that second time point to estimate the state of the diagnostic target sensor.

[0048] In the first embodiment, the sensor state estimation unit 56 compares the estimated value of the detection result of the third sensor device 20c at the second time point, which is the detection result of the diagnostic target sensor at the second time point, with the actual detection result (measured value) of the third sensor device 20c represented by the third sensor data acquired at the second time point, which is the detection result of the diagnostic target sensor acquired at the second time point. If the difference between the estimated value of the detection result of the third sensor device 20c and the actual detection result (measured value) of the third sensor device 20c deviates from a predetermined allowable range, the sensor state estimation unit 56 estimates that the state of the third sensor device 20c is abnormal. This allowable range can be determined by the developer of the diagnostic system 10 based on the insights of the developer and experiments using the diagnostic system 10 (e.g., experiments using a normal third sensor device 20c and an abnormal third sensor device 20c, respectively).

[0049] The sensor state estimation unit 56 saves diagnostic information, including the estimation result of the state of the sensor to be diagnosed and the estimated date and time, to the diagnostic information storage unit 48. In response to a request from an external source, the diagnostic information providing unit 58 sends the diagnostic information stored in the diagnostic information storage unit 48 to an external device (not shown) (e.g., a maintenance worker's terminal), or the diagnostic information providing unit 58 periodically sends the diagnostic information stored in the diagnostic information storage unit 48 to an external device (not shown) (e.g., a maintenance worker's terminal).

[0050] The operation of the diagnostic system 10 based on the above structure in the first embodiment is explained.

[0051] Figure 4This is a flowchart illustrating the operation of the diagnostic system 10 according to the first embodiment. Here, the first time point and the second time point are assumed to be the same time point for explanation.

[0052] The detection unit 30 of the first sensor device 20a periodically detects vibrations at the sensor mounting position in the robotic arm 12. The antenna 38 of the first sensor device 20a sends first sensor data representing the detection result of the detection unit 30 to the diagnostic device 22. In parallel, the second sensor device 20b also detects vibrations at the sensor mounting position and sends second sensor data representing its detection result to the diagnostic device 22. Similarly, the third sensor device 20c also detects vibrations at the sensor mounting position and sends third sensor data representing its detection result to the diagnostic device 22. The detection result acquisition unit 52 of the diagnostic device 22 acquires the first sensor data, second sensor data, and third sensor data periodically sent from the first sensor device 20a, second sensor device 20b, and third sensor device 20c (S10).

[0053] At a certain time point (first time point), the operation condition information acquisition unit 50 of the diagnostic device 22 acquires the first control signal, the second control signal, and the third control signal sent by the robotic arm control device 18 to the robotic arm 12 (S12).

[0054] The detection result estimation unit 54 of the diagnostic device 22 inputs the operation conditions of the first movable part 14a, the operation conditions of the second movable part 14b, and the operation conditions of the third movable part 14c, as indicated by the first control signal acquired by the operation condition information acquisition unit 50 at the first time point, the detection results represented by the first sensor data acquired by the detection result acquisition unit 52 at the first time point, and the detection results represented by the second sensor data, into the simulation model. The detection result estimation unit 54 acquires the detection result (estimated value) of the diagnostic target sensor (third sensor device 20c) at the second time point (here, the same time point as the first time point) estimated by the simulation model (S14).

[0055] The sensor state estimation unit 56 of the diagnostic device 22 compares the detection result (estimated value) of the third sensor device 20c obtained from the simulation model with the detection result (actual measured value) represented by the third sensor data acquired by the detection result acquisition unit 52 at the first time point to estimate the state of the sensor to be diagnosed, namely the third sensor device 20c (S16). The sensor state estimation unit 56 saves the diagnostic information representing the estimation result of the state of the third sensor device 20c to the diagnostic information storage unit 48. The diagnostic information providing unit 58 of the diagnostic device 22 provides the diagnostic information related to the third sensor device 20c stored in the diagnostic information storage unit 48 to an external device (S18).

[0056] According to the diagnostic system 10 of the first embodiment, the detection results of the sensor device 20, which is the object of diagnosis, provided on the robotic arm 12, are estimated using parameters of the operating conditions of the movable part 14 including the robotic arm 12. This improves the estimation accuracy. Specifically, the diagnostic system 10 can obtain an estimate that approximates the detection results of the sensor device 20 under normal conditions. Furthermore, this improves the accuracy of estimating the state of the sensor device 20.

[0057] <Second Embodiment>

[0058] The second embodiment of the present invention will be described focusing on the points that differ from the embodiments described above, and descriptions of common points will be appropriately omitted. The features of the second embodiment can, of course, be arbitrarily combined with the features of the embodiments and variations described above. Elements in the second embodiment that are the same as or correspond to elements in the embodiments described above will be appropriately labeled with the same reference numerals for description.

[0059] The structure of the diagnostic system 10 in the second embodiment and Figure 1 The diagnostic system 10 of the first embodiment shown has the same structure. The diagnostic system 10 of the second embodiment also diagnoses the status of the sensor device 20 installed on the robotic arm 12 in the same way as the diagnostic system 10 of the first embodiment. Figure 5 This is a block diagram showing the functional blocks of the diagnostic device 22 of the second embodiment. In addition to the functional blocks of the diagnostic device 22 of the first embodiment, the diagnostic device 22 of the second embodiment also includes a degradation estimation unit 60 and an update unit 62.

[0060] In addition to storing the simulation model data described in the first embodiment, the model storage unit 46 also stores data for a degradation estimation model used to estimate the degree of degradation of the robotic arm 12. The degradation estimation model is a mathematical model that takes multiple sensor data (first sensor data, second sensor data, and third sensor data) representing the detection results of multiple sensor devices 20 as input to estimate the degree of degradation of the robotic arm 12. The first sensor data, second sensor data, and third sensor data in the second embodiment can be, like in the first embodiment, information related to vibration at the sensor placement location.

[0061] For example, the degradation estimation model can be a regression equation with the detection results of the first sensor device 20a (represented by the first sensor data), the detection results of the second sensor device 20b (represented by the second sensor data), and the detection results of the third sensor device 20c (represented by the third sensor data) as explanatory variables, and the degradation index representing the degree of degradation of the robotic arm 12 as the target variable. Alternatively, when constructing the degradation estimation model, the actual values ​​of the detection results of the first sensor device 20a, the second sensor device 20b, the third sensor device 20c, and the degree of degradation of the robotic arm 12 can be collected as sample data. Furthermore, the coefficients of each explanatory variable can be determined by performing multiple regression analysis based on multiple sample data, thereby generating the data for the degradation estimation model.

[0062] In addition, the model storage unit 46 stores data of multiple simulation models corresponding to the degree of degradation of the robotic arm 12. These multiple simulation models may also be generated based on sample data collected from robotic arms 12 with varying degrees of degradation (as described in the first embodiment). The model storage unit 46 may also store multiple degradation index values ​​associated with simulation models suitable for the degree of degradation represented by each degradation index value.

[0063] The degradation estimation unit 60 estimates the degree of degradation of the robotic arm 12 based on the detection results of the multiple sensor devices 20. Specifically, the degradation estimation unit 60 estimates the degree of degradation of the robotic arm 12 by reading the data of the degradation estimation model stored in the model storage unit 46 and inputting the detection results of the multiple sensor devices 20 sent from the multiple sensor devices 20 into the degradation estimation model.

[0064] The updating unit 62 updates the simulation model used by the detection result estimation unit 54 based on the degree of degradation of the robotic arm 12 estimated by the degradation estimation unit 60. In the second embodiment, data of a simulation model suitable for the degree of degradation of the robotic arm 12 estimated by the degradation estimation unit 60 is read from a plurality of simulation models stored in the model storage unit 46, and the read simulation model data is transmitted to the detection result estimation unit 54.

[0065] The detection result estimation unit 54 uses the simulation model updated by the update unit 62, that is, the simulation model transmitted from the update unit 62 in the second embodiment that is suitable for the degree of degradation of the robotic arm 12, to estimate the detection result of the diagnostic object sensor.

[0066] When the degradation level of the robotic arm 12 exceeds a predetermined threshold, the sensor state estimation unit 56 estimates the state of the sensor to be diagnosed. Furthermore, assuming the degradation level of the robotic arm 12 exceeds the predetermined threshold, the detection result estimation unit 54 estimates the detection result of the sensor to be diagnosed; thus, the sensor state estimation unit 56 estimates the state of the sensor to be diagnosed. This threshold can be determined based on the insights of the developer of the diagnostic system 10, experiments using the diagnostic system 10, or the probability of sensor failure at each degradation level of the robotic arm 12, etc.

[0067] The operation of the diagnostic system 10 of the second embodiment of the above structure will be explained.

[0068] Figure 6 This is a flowchart illustrating the operation of the diagnostic system 10 according to the second embodiment. Similar to the first embodiment, the detection result acquisition unit 52 of the diagnostic device 22 acquires first sensor data, second sensor data, and third sensor data periodically transmitted from the first sensor device 20a, the second sensor device 20b, and the third sensor device 20c (S20). The degradation estimation unit 60 of the diagnostic device 22 inputs the detection results of the first sensor device 20a represented by the first sensor data, the detection results of the second sensor device 20b represented by the second sensor data, and the detection results of the third sensor device 20c represented by the third sensor data into the degradation estimation model. The degradation estimation unit 60 acquires a degradation index value derived using the degradation estimation model, representing the degree of degradation of the robotic arm 12 (S22).

[0069] At a certain time point (first time point), the operation condition information acquisition unit 50 of the diagnostic device 22 acquires the first control signal, the second control signal, and the third control signal sent by the robotic arm control device 18 to the robotic arm 12 from the robotic arm control device 18 (S24). If the degradation level of the robotic arm 12 is above a predetermined threshold (S26: "Yes"), the update unit 62 reads the data of the simulation model corresponding to the degradation level of the robotic arm 12 from the model storage unit 46, in other words, the data of the simulation model associated with the degradation index value acquired in S22, and transmits the data to the detection result estimation unit 54 (S28).

[0070] The subsequent processing of S30 to S34 and Figure 4The processing of S14 to S18 in the diagnostic system 10 of the first embodiment shown is the same, so the description is omitted. If the degree of degradation of the robotic arm 12 is less than the above-mentioned threshold (S26: "No"), the processing after S28 is skipped. According to the diagnostic system 10 of the second embodiment, by using a simulation model corresponding to the degree of degradation of the robotic arm 12, the estimation accuracy of the detection results of the sensor of the diagnostic object can be improved. In addition, according to the diagnostic system 10, the abnormality of the sensor can be detected efficiently by estimating the state of the sensor of the diagnostic object when the degradation of the robotic arm 12 has increased to a certain extent.

[0071] The present invention has been described above based on the first and second embodiments. It will be understood by those skilled in the art that the embodiments are illustrative, and the combination of constituent elements and processing steps described in the embodiments can have various modifications, and such modifications are also within the scope of the present invention.

[0072] As a first variation, a variation related to the second embodiment will be described.

[0073] The multiple sensor devices 20 installed on the robotic arm 12 are used to detect vibrations at each sensor installation location, similar to those in the first embodiment. The degradation estimation unit 60 of the diagnostic device 22 compares the vibration convergence time at the sensor installation location calculated based on the detection results of the multiple sensor devices 20 with the vibration convergence time estimated based on the motion condition information to estimate the degree of degradation of the robotic arm 12.

[0074] Specifically, the degradation estimation model stored in the model storage unit 46 can also accept the input of detection results from each of the multiple sensor devices 20 and derive the vibration convergence time (measured value) for the installation position of each sensor device 20. The vibration convergence time can also be described as the duration of vibration, and this vibration convergence time can be derived based on known benchmarks. For example, the vibration convergence time is the time from detecting vibration of a magnitude above a predetermined first threshold to the point where vibration below the first threshold and above a second threshold can no longer be detected. Furthermore, the degradation estimation model can also accept the input of the operating condition information represented by each of the first, second, and third control signals, and calculate the vibration convergence time (estimated value) for the installation position of each sensor device 20 based on a predetermined evaluation function. The evaluation function outputs the vibration convergence time of the sensor installation position, pre-associated with each operating condition, according to the operating conditions represented by the operating condition information.

[0075] The degradation estimation model can also compare the vibration convergence time (measured value) with the vibration convergence time (estimated value) at each of the three sensor devices (first sensor device 20a, second sensor device 20b, and third sensor device 20c) and estimate the degree of degradation of the robotic arm 12 based on the comparison results. For example, if the vibration convergence time (measured value) is greater than the vibration convergence time (estimated value) at zero or one sensor device location, the degradation degree can be estimated as low. Alternatively, if the vibration convergence time (measured value) is greater than the vibration convergence time (estimated value) at two sensor device locations, the degradation degree can be estimated as moderate. Furthermore, if the vibration convergence time (measured value) is greater than the vibration convergence time (estimated value) at all three sensor device locations, the degradation degree can be estimated as high.

[0076] The degradation estimation unit 60 of the diagnostic device 22 inputs vibration information of the sensor installation position represented by each sensor data from the plurality of sensor data sent from the plurality of sensor devices 20 into the degradation estimation model. Additionally, the degradation estimation unit 60 inputs the operating conditions represented by each of the first, second, and third control signals sent from the robotic arm control device 18 into the degradation estimation model. Furthermore, the degradation estimation unit 60 acquires an index value representing the degree of degradation of the robotic arm 12 estimated by the degradation estimation model. Subsequent processing is the same as in the second embodiment, and therefore description is omitted. Based on the structure of this modified example, the degree of degradation of the robotic arm 12 can be estimated with high accuracy based on the vibration convergence time of each sensor installation position.

[0077] As a second variation, a variation relating to both the first and second embodiments will be described.

[0078] In the first and second embodiments, sensor devices 20 are provided between the plurality of movable parts 14 (links 16) of the robotic arm 12. However, as a variation, sensor devices 20 may also be provided in the plurality of movable parts 14.

[0079] As a third variation, other variations relating to both the first and second embodiments will be described.

[0080] In the first and second embodiments, the diagnostic target sensor is designated as the third sensor device 20c. However, any one or more of the first sensor device 20a, second sensor device 20b, and third sensor device 20c can also be designated as the diagnostic target sensor. In this case, the model storage unit 46 of the diagnostic device 22 can also store simulation models, different for each diagnostic target sensor, used to estimate the state of each diagnostic target sensor. The detection result estimation unit 54 and the sensor state estimation unit 56 can also use simulation models corresponding to one or more diagnostic target sensors respectively to estimate the state of each diagnostic target sensor.

[0081] The sensor device 20 in the above embodiment is designed as a sign sensor device, but as a variation, the sensor device 20 may not be a sign, but may be a thin sheet or coin-shaped sensor device that can easily stick to the object.

[0082] Regarding the parts in the embodiments disclosed in this specification that have multiple functions distributedly, some or all of the multiple functions can also be integrated. Conversely, a part that has multiple functions integratedly can be distributed in a manner where some or all of the multiple functions are distributed. Whether the functions are integrated or distributed, they can be configured in a way that achieves the purpose of the invention.

[0083] Any combination of the above-described embodiments and modifications is also useful as an implementation of the present invention. New embodiments resulting from such combinations combine the effects of both the combined embodiments and modifications. Furthermore, those skilled in the art will understand that the functions to be performed by the constituent elements described in the claims are achieved by the individual elements shown in the embodiments and modifications, or by their collaboration.

[0084] Furthermore, the techniques described in the embodiments and variations can also be determined in the following ways.

[0085] [Project 1]

[0086] A diagnostic system, comprising:

[0087] An action condition information acquisition unit acquires action condition information that can identify the action conditions of the device, wherein the device has a movable part that is driven by a driving force.

[0088] The detection result acquisition unit acquires the detection results from multiple sensors installed on the device;

[0089] The detection result estimation unit estimates the detection result of a specific sensor among the multiple sensors at a second time point by inputting the action condition information acquired at a first time point and the detection results of the multiple sensors acquired at the first time point into the simulation model; and

[0090] The sensor state estimation unit compares the estimated detection result of the specific sensor at the second time point with the detection result of the specific sensor among the plurality of sensors acquired at the second time point to estimate the state of the specific sensor.

[0091] The second time point can be either the same as the first time point or a different time point. The sensor state estimation unit can also be described as a sensor anomaly detection unit that detects an abnormal sensor state.

[0092] According to this diagnostic system, by taking into account the operating conditions of the equipment to estimate the detection results of a specific sensor, the estimation accuracy of the sensor's detection results can be improved, thereby improving the estimation accuracy of the sensor's state.

[0093] [Project 2]

[0094] According to the diagnostic system described in Project 1, among which,

[0095] The device has multiple movable parts.

[0096] The detection result estimation unit estimates the detection result of the specific sensor at a second time point by inputting the operating conditions of the plurality of movable parts into the simulation model.

[0097] According to this diagnostic system, when the device has multiple movable parts, the estimation accuracy of the sensor's detection results can be further improved by estimating the detection results of a specific sensor based on the operating conditions of these multiple movable parts.

[0098] [Project 3]

[0099] According to the diagnostic system described in item 1 or 2, among which,

[0100] The detection result estimation unit estimates the detection result of the specific sensor at a second time point by inputting the detection results of the sensors other than the specific sensor from the plurality of sensors into the simulation model.

[0101] According to this diagnostic system, by excluding the detection results of specific sensors that may have outliers, the system can accurately estimate the situation when the sensor is in an abnormal state.

[0102] [Project 4]

[0103] According to the diagnostic system described in any one of items 1 to 3, among which,

[0104] A degradation estimation unit estimates the degree of degradation of the device based on the detection results of the plurality of sensors; and

[0105] The update unit updates the simulation model based on the degree of degradation of the device estimated by the degradation estimation unit.

[0106] According to this diagnostic system, by using a simulation model corresponding to the degree of equipment degradation, the estimation accuracy of detection results for specific sensors can be improved.

[0107] [Project 5]

[0108] According to the diagnostic system described in Project 4, among which,

[0109] When the degree of degradation of the device estimated by the degradation estimation unit is above a predetermined threshold, the sensor state estimation unit estimates the state of the specific sensor.

[0110] According to this diagnostic system, anomalies of a particular sensor can be detected efficiently by estimating the state of that sensor when the equipment deteriorates further.

[0111] [Project 6]

[0112] According to the diagnostic system described in item 4 or 5, among which,

[0113] The multiple sensors detect vibrations at their respective installation locations.

[0114] The degradation estimation unit compares the vibration convergence time at the sensor location calculated based on the detection results of the multiple sensors with the vibration convergence time estimated based on the operating condition information to estimate the degree of degradation of the device.

[0115] According to this diagnostic system, the degree of equipment degradation can be estimated with high accuracy based on the vibration convergence time at the sensor setting location.

[0116] Industrial availability

[0117] The technology disclosed herein can be applied to diagnostic systems.

[0118] Explanation of reference numerals in the attached figures

[0119] 10: Diagnostic system; 12: Robotic arm; 14: Movable part; 20: Sensor device; 22: Diagnostic device; 50: Action condition information acquisition unit; 52: Detection result acquisition unit; 54: Detection result estimation unit; 56: Sensor state estimation unit; 60: Deterioration estimation unit; 62: Update unit.

Claims

1. A diagnostic system, comprising: An action condition information acquisition unit acquires action condition information that can identify the action conditions of the device, wherein the device has a movable part that is driven by a driving force. The detection result acquisition unit acquires the detection results from multiple sensors installed on the device; The detection result estimation unit estimates the detection result of a specific sensor among the multiple sensors at a second time point by inputting the action condition information obtained at a first time point and the detection results of the multiple sensors obtained at the first time point into the simulation model. The sensor state estimation unit compares the estimated detection result of the specific sensor at the second time point with the detection result of the specific sensor among the plurality of sensors acquired at the second time point to estimate the state of the specific sensor. The degradation estimation unit estimates the degree of degradation of the device based on the detection results of the plurality of sensors; as well as The update unit updates the simulation model based on the degree of degradation of the device estimated by the degradation estimation unit.

2. The diagnostic system according to claim 1, wherein, The device has multiple movable parts. The detection result estimation unit estimates the detection result of the specific sensor at a second time point by inputting the operating conditions of the plurality of movable parts into the simulation model.

3. The diagnostic system according to claim 1 or 2, wherein, The detection result estimation unit estimates the detection result of the specific sensor at a second time point by inputting the detection results of the sensors other than the specific sensor from the plurality of sensors into the simulation model.

4. The diagnostic system according to claim 1, wherein, When the degree of degradation of the device estimated by the degradation estimation unit is above a predetermined threshold, the sensor state estimation unit estimates the state of the specific sensor.

5. The diagnostic system according to claim 1 or 4, wherein, The multiple sensors detect vibrations at their respective installation locations. The degradation estimation unit compares the vibration convergence time at the sensor location calculated based on the detection results of the multiple sensors with the vibration convergence time estimated based on the operating condition information to estimate the degree of degradation of the device.

Images